550
BIOCHEMISTRY
H A R A , CHEW, A h D SONEVBERCJ
Recombination of the Biologically Active Peptides from a Tryptic Digest of Bovine Growth Hormone? Kenji Hara,r Chen-Jung Hsu Chen,§ and Martin Sonenberg*
ABSTRACT:
Two bovine growth hormone peptides containing residues 96 to 133 in one and a larger peptide corresponding to residues 1 to 95 and 15 1 to 19 1 linked by a disulfide bond have been recombined. Recombination was performed in 1 .O N acetic acid without and with 8 M urea and the recombined peptides were purified by gel filtration. Evidence of recombination of the peptides has been established by gel filtration, polyacrylamide gel disc electrophoresis, amino acid composition, ultraviolet absorption, circular dichroism, and intrinsic
fluorescence. The molar growth promoting activity of the recombined peptides measured by tibial width assay in hypophysectomized rats was 10% that of the native hormone and greater than the sum (1%) of the growth promoting activities of the individual peptides. These studies suggest that some contribution of the larger peptide is necessary for more complete expression of the growth promoting activity of the bovine growth peptide corresponding to residues 96 to 133.
F r o m tryptic digests of bGH (TbGH)' we have isolated a single component, TbGH-d, as demonstrated by analytical disc electrophoresis and sedimentation equilibrium (Sonenberg et al., 1968). We have reported (Yamasaki et al., 1970) fractionation of TbGH-d by gel filtration in 50% acetic acid and have isolated two components, bGH 1-95 and 151-191, a peptide of 16 000 molecular weight. and bGH 96-133, a peptide of 5000 molecular weight (Figure 1). bGH 1-95and 151-191 andbGH96-133areactiveinthe weight gain and tibial width test in hypophysectomized rats, but the level of activity was much less than that of native bGH from which these peptides were produced (Yamasaki et al.. 1970; Hara and Sonenberg, 1977). bGH 1-95 and 151-191 was much less active in growth promoting assays than bGH 96-133 (Yamasaki et al., 1970; Hara and Sonenberg, 1977). Both peptides were also active in vitro in stimulating glucose uptake and conversion to carbon dioxide, lipolysis, glycerol release, and histidine incorporation into protein in isolated adipose tissue of hypophysectomized rats (Swislocki et al.. 1970). In addition, bGH 96-1 33 was found to have metabolic activity in humans similar to h G H (Sonenberg et al., 1972, 1973; Levineet al., 1973). The amino acid sequence of bGH 96- 133 has been determined (Yamasaki et al., 1972, 1975). It consists of 38 amino
acid residues and is homologous with region 96-133 of the bGH molecule (Wallis, 1973; Santomt et al., 1973; GrAf and Li, 1974a). Because of the good homology in the amino acid sequence between bGH 96- 133 and a corresponding peptide occurring in hGH, human chorionic somatomammotropin, and ovine growth hormone, b G H 96-1 33 may represent some of the amino acid sequence responsible for its metabolic activities (Swislocki et al., 1970; Yamasaki et al., 1972, 1975). TbGH-d, the parent component from which bGH 1-95 and 151 - 19 1 and bGH 96- I33 were derived, retained full activity compared with bGH (Yamasaki et al., 1970; Sonenberg et al., 1972). bGH 96-133 may have reduced growth promoting activity because of a secondary structure and tertiary structure significantly different from the native hormone from which it was derived. This could result in impaired affinity for the target cell receptor for growth hormone and changes in its ability to stimulate a biological response. These considerations form the basis of separate studies (submitted). The possibility remained that the low growth promoting potency of the smaller peptide (bGH 96-133) was the consequence of the loss of some contribution of the larger peptide (bGH 1-95 and 151-191) to the size and conformation of the smaller peptide. In order to evaluate the contribution of the larger peptide to the growth promoting activity of the smaller peptide, an attempt has been made to recombine the biologically active smaller peptide with the larger peptide. In this paper, only growth promoting activity is considered. Metabolic and immunologic activities have previously been reported (Swislocki et al.. 1970; Sonenberg et al., 1973). Preliminary reports of other recombination studies from this laboratory have appeared (Sonenberg and Beychok. 1971; Sonenberg et a]., 1972). Recombination of fragments of hGH with nearly full restoration of biological activity has also been reported (Li and Bewley, 1976; Li et al., 1976, 1977).
~~
From the Memorial Sloan-Kettering Cancer Center and Cornell University. Graduate School of Medical Sciences, New York, New York 1002 1 . Received April 18, 1977. Supported in part by Grants CA-08748 of the National Cancer Institute and BC-147 of the American Cancer Societ) t Present address: Kao Tokyo Research Institute, 2-1-3. Bunka. S u m ida-ku. Tokyo. 131, Japan. Present address: Department of Pathology, Yale University School of Medicine. New Haven, Conn. 06510. Abbreviations used: bGH, bovine pituitary growth hormone; TbGH, tryptic digest of bovine growth hormone; TbGH-d, homogeneous component of a tryptic digest of bovine growth hormone prepared by Yamasaki et al. (1970): TbGH-d (HoAC), TbGH-d exposed to 50% acetic acid; TbGH-x, homogeneous component of tryptic digest of bovine growth hormone prepared by the methods of Yamasaki et a]. (1970) for this study; bGH 1 -95 and 15 I - I9 I , peptide of bovine growth hormone corresponding to residues 1 to 95 and I 5 1 to 191 linked by a disulfide bond; bGH 96-1 33, peptide of bovine growth hormone corresponding to residues 96 to 133: bGH 134-1 50, peptide of bovine growth hormone corresponding to residues 134 to 150; hGH, human pituitary growth hormone; oGH, ovine growth hormone; NaDodSO1, sodium dodecyl sulfate; dansyl, Dns, I dinicthylamino-5-naphthalenesulfonyl:CD, circular dichroism. +
'
Experimental Procedures Materials. bGH was prepared by the methods previously reported (Dellacha and Sonenberg, 1964; Free and Sonenberg, 1966). Trypsin and soybean trypsin inhibitor were obtained from Worthington Biochemical Corp., Freehold, N.J. Urea (Ultrapure grade, no copper, iron, lead, or cyanide) was a product of Schwarz/ Mann, Orangeburg, N . Y . 1 -Dimethylamino-5-naphthalenesulfonyl chloride (Dns) ( 1 0% solution 1978 American Chemical Society
RECOMBINATION OF GROWTH HORMONE PEPTIDES
in acetone) was obtained from Pierce Chemical Co., Rockford, Ill. Polyamide layer sheets for the separation of Dns-amino acids were obtained from Gallard-Schlesinger Chemical Mfg. Corp., Carle Place, N.Y. Other chemicals used in this study were reagent grade. Preparation and Fractionation of Tryptic Digests of Bovine Growth Hormone. Tryptic digests of b G H were prepared by the methods previously reported (Sonenberg et al., 1965, 1968; Nadler et al., 1968; Yamasaki et al., 1970). The single component of the tryptic digest of b G H (TbGH-x) was isolated from the tryptic digests of bGH by ion-exchange chromatography on a DEAE-cellulose column by the methods of Yamasaki et al. (1970). TbGH-x, with which subsequent studies were done, appeared in fractions eluted with 0.03 and 0.05 M ammonium bicarbonate buffer. It had the same electrophoretic mobility previously designated the d component of the tryptic digests of bGH (TbGH-d). For further purification of TbGH-x, the same chromatographic system was employed. The component which was homogeneous by analytical disc electrophoresis was used for the isolation of the individual peptides. Isolation and Characterization of the Biologically Active Peptides. The biologically active peptides were prepared from a single component, TbGH-x, of tryptic digests of bGH, dissociated by 5G?h acetic acid as previously described (Yamasaki et al., 1970). In addition, gel filtration of the single component (TbGH-x) of T b G H was carried out on a Sephadex G-75 column with 0.1 M carbonate buffer (pH 9.5) containing 8 M urea. The proteins were identified in four fractions. From 300 mg of TbGH-x, the yield was 70 mg of the most rapidly eluted, 1 2 0 m g o f b G H 1-95and 151-191,55mgofbGH96-133and 20 mg of bGH 134- 150. The peptide fractions were pooled and applied on a Sephadex G-25 or Bio-Gel P-2 column to remove urea with 0.01 M ammonium bicarbonate buffer (pH 8.4) and lyophilized. The three most rapidly eluted fractions were further purified by repeating the gel filtration in 0.1 M carbonate buffer (pH 9.5) containing 8 M urea. The purity of these peptides was established by disc electrophoresis, amino-terminal amino acids, amino acid composition, and spectrofluorescence. From fluorescence spectra of b G H 96-133, only tyrosine fluorescence was noted whereas bGH 1-95 and 15 1 - 19 1 had evidence of tryptophan fluorescence. The molecular weight of bGH 96-133 was determined to be 5000 by using 10%cross-linked polyacrylamide gel NaDodS04 disc electrophoresis. From amino acid analysis, ultraviolet absorption, fluorescence, and circular dichroism spectra, it was concluded that the 5000 molecular weight peptide originated in residues 96-133 (Figure 1) of the b G H molecule (Wallis, 1973; SantomC et al., 1973; Gr6f and Li, 1974a). The largest peptide consisted of the amino-terminal portion (1-95 of the sequence of bGH) and carboxyl-terminal position (1 5 1- 19 1 of the sequence of bGH) which are linked by a disulfide bond (Figure 1). The smallest peptide was separated and urea was removed by gel filtration through Bio-Gel P-2. The purified peptide was subjected to paper chromatography in the 1-butanol-pyridine-acetic acid-water (15:10:3:12, v/v) system and one homogeneous peptide ( R j 0.29) was obtained and consisted of residues 134-150 (Figure 1) of the b G H molecule (Wallis, 1973; SantomC et al., 1973; Grdf and Li, 1974a). Only limited studies were performed with this peptide. Recombination of Peptides. For the recombination of peptides derived from tryptic digests of bGH, 32 mg of bGH 1-95 and 15 1 - 19 1 and 10 mg of b G H 96- 133 were dissolved separately to make a 1 .O% protein solution in 1.0 N acetic acid. These were mixed and stirred continuously for 5 h a t room
VOL.
17,
NO.
3, 1978
551
I : Schematic representation of the components of TbGH-x. bGH 1-95 and 151-191 (-), bGH 96-133 (-), and bGH 134-150 (- - -) modified from Chen and Sonenberg (1977). FIGURE
temperature. The reaction mixture was freeze dried directly. To purify the recombined complex, the lyophilized reaction mixture was applied to a Sephadex G-75 column and eluted with 1.O N acetic acid. Recombination in urea solution was performed as follows; 3 2 m g o f b G H 1-95and 151-191 a n d l O m g o f b G H 9 6 - 1 3 3 were dissolved separately in 3.2 mL and 1.O mL of 1.O N acetic acid solution containing 8 M urea. Both peptide solutions were mixed and stirred constantly for 5 h at room temperature. The reaction mixture was transferred into a Visking tube and dialyzed against 1.O N acetic acid solution containing 8 M urea for 24 h. The outer solution was then changed to 1.O N acetic acid solution containing sequentially 6 M urea, 4 M urea, 2 M urea, and no urea. The recombined complex was recovered by freeze-drying. Purification of Recombined Peptides. The lyophilized crude products of recombined peptides were applied on a column of Sephadex G-75 equilibrated with 1.0 N acetic acid. The column was developed with the same solution. The pooled fractions were lyophilized directly. For characterization of the recombined peptides, fractions were further purified by repeating the gel filtration in 1.O N acetic acid. Disc Electrophoresis with and without Sodium Dodecyl Sulfate. Disc electrophoresis in polyacrylamide gels was carried out by the methods of Ornstein (1964), Davis (1964), and Reisfeld et al. (1962). Electrophoresis was performed a t p H 9.5 and/or p H 4.3 in 7.5% polyacrylamide gel. Protein components were detected by staining with Amido Black 10B, followed by removing unbound dye with several changes of 7% acetic acid. Polyacrylamide gel disc electrophoresis in NaDodS04 was carried out by the methods of Shapiro et al. (1 967), and Weber and Osborn (1969). The marker proteins (bacitracin, cytochrome c , chymotrypsinogen A, ovalbumin, albumin) for the determination of molecular weight were purchased from Schwarz/Mann, Orangeburg, N.Y. Amino Acid Analysis and Amino- Terminal Amino Acid Analysis. Protein or peptide samples were hydrolyzed in vacuo with 6.0 N HCI a t 110 O C for 24 h. Hydrolysates were analyzed on an automatic amino acid analyzer Beckman Model 121 according to Spackman et al. (1958). Amino-terminal amino acid analysis was performed by the method of Gray and Hartley (1963). An aliquot of protein or peptide sample was reacted with dansyl chloride. The resulting Dns-peptide was then hydrolyzed in 6.0 N HC1 a t 106 O C for 16 h in vacuo. Dns-amino acids were separated and identified by two-dimensional polyamide layer chromatography (Woods and Wang, 1967). Development in a covered jar by ascending solvent systems was employed: (I) water-90% formic acid
552
BIOCHEMISTRY
HARA, CHEN, AND SONENBERO
TABLE I:
Amino Acid Composition of Rewmbined Peptides.',b
Amino acid
. 8
ASP Thr
i
'0
20
40 GO Tube number
80
Glu Pro GlY Ala 1/2-Cystine Val Met
100
2 Gel filtration of (A) rccombined peptides, bGH 1-95 and 151-191 andbCH96-133(30mg)and(B) bGH (30mg)anSephadex G-75 column (2.5 X 43.0 cm) in 1.0 N acetic acid; fraction volume. 2.1 mL. FIGURE
0
bCH 1-95 and I5I-19Ic
bGH I-95and 96-133d
+
1-133'
bGH96-133
bGH96-133
151-191
14.1 9.7 10.4 22.0 6.6 9.1 12.1 3.3 6.4 2.4
13.8 9.3 10.3 22.0 6.3 9.3 11.4 3.4 7.3 2.9 5.6 23.3 5.0 11.9 9.5 3.4 14.2
13 IO 13 22 6 9 14 4 6 3 6 26 5 12 9 3 I2
+
lle
6.1
Le" Tyr
23.1 4.5 9.9 8.8 2.4 12.6
Phe LYS His Ara
~~
+
a Composition in molar ratio. Thr and Ser values arc uncorrected. Viffccrcnt 10th of pptidm have becn cmplo3cd in each remmbinatiun. Rcwnibincd in I .O \ acetic wid sonwining 8 M urea. Rmmbincd in I.0N acctic acid. e Rcsiduc positions in the bGH structure (Wallis. 1973; Santnm2 et al.. 1973; G r i f :Ind Li. 1974a.b) Knt dcter-
I'
mincd
bGH
TbGH-x
Recombined peptides of bGH
FIGURE 3 Polyacrylamide gel disc electrophoresis of bCH, TbGH-x, and recombined bGH 1-95 and 151-191 and bGH 96-133.
(200:3, v/v); (11) benzene-glacial acetic acid (9:l, v/v) (Woods and Wang, 1969). Ultraviolet Absorption Spectra. Ultraviolet absorption spectra were measured with a Cary Model 11 recording spectrophotometer. Cylindrical quartz tandem cells (IO-mm path length) were used so that both peptide solutions were in the light path, either separated by a quartz partition, or mixed in the same compartment. The protein concentrations were 0.3 mg/mL for bGH 96-133.0.96 mg/mL for bGH 1-95 and 151-191, and 1.26 mg/mL for separated and mixed peptides in 0.1 N acetic acid solution. Ultraviolet absorption spectra were also measured in 0.1 M sodium carbonate buffer (pH 9.5) to compare with bGH. Circular Dichroism Spectra. Circular dichroism determinations were made in the far ultraviolet range (250-195 nm) with a Cary Model 60 recording spectropolarimeter with circular dichroism attachment 6002. Cylindrical quartz tandem cells (IO-mm path length) were used so that solutions of bGH 1-95 and 151-191 and b G H 96-133 were both in the light path, either separated by a quartz partition, or mixed in the same compartment. Aliquots of the protein solutions were taken for Lowry-Folin test (Lowry et al., 1951) todetermine the protein concentration for each experiment. Fluorescence Spectra of the Recombined Peptides. Fluorescence measurements were made on a Cary 50-026-900 differential spectrofluorimeter equipped with a Rhodamine B quantum detector located within the sample compartment and employing front surface illumination, where the exciting and emitting beams subtend an angle of 23O. Other details with respect to the spectrofluorimeter were as described elsewhere (Sonenberg, 1971). Measurements were made with the exci-
tation monochromator set at 270 nm at 25 "C. The concentration of peptides was 0.1% in 0.1 M sodium carbonate buffer (pH 9.5). Biological Response. The growth responses of the individual and the recombined peptides were determined in hypophysectomized rats by the 4 day tibial width response test (Geschwind and Li, 1955). Five rats per dose were employed. Results
Purification of Recombined Peptides. The lyophilized products obtained by recombination in 1.ON acetic acid were purified by Sephadex G-75 column chromatography from peptidesbGH I-95and 151-191 andbGH96-133previously equilibrated with 1 .O N acetic acid and eluted with the same solution (Figure 2A). Onemain fraction (fraction b) and two minor fractions were obtained. Fraction a was eluted at the same volume as blue dextran. I t was found to contain aggregated material and was not investigated further. Similarly, the other minor fraction, fraction c, was not further characterized. bGH on a calibrated column under the same conditions was eluted at the same volume as fraction b (Figure 1 B). The main fraction b was purified further by rechromatography on Sephadex G-75 in 1.0 N acetic acid and a single peak was noted. Analytical electrophoresis on polyacrylamide gel indicated that the recombined complex was homogeneous. The recombined complex had a greater cathodal mobility than bGH and TbGH-x (Figure 3). Amino Acid Composition. Comparison of the amino acid compositions reveals almost identical compositions between the recombined peptides of bGH and bGH without residues 134-150 (Table I). Spectroscopic Characterization of Recombined Peptides. Typical absorption spectra (not shown) of recombined peptides as established above were compared with that of peptides bGH 1-95 and 151-191 a n d b c H 96-133. The recombined peptides
RECOMBINATION OF GROWTH HORMONE PEPTIDES
VOL.
1 7 , NO. 3, 1978
553
I
I
iI
J
4
-151-10/
\ \
15
190
200
210
220
230
240
- ,/---'
250
Wavelength (nml
4: Circular dichroism spectra of recombined bGH 1-95 and 151-191 and bGH 96-133 (- - -), unmixed equimolar concentrations of bGH 1-95 and 151-191 and b C H 96-133 in separate compartments -), bGH 1-95 and 151-191 (-),and bGH of a quartz tandem cell 96-133 (- - -) in 0.1 N acetic acid; [ e ] , deg cm2 dmol-'. FIGURE
(-e
TABLE 11: Secondary Structure Analysis
of the Recombined
Peotides.
Protein
pH
Helixa
pa
b G H 1-95 and 151-191, b G H 96-133, complex b G H 1-95 and 151-191 b G H 96-133 (calcd) TbGH-d T b G H - d (HoAC) b C H 1-95 and 151-191, b G H 96-133, complex b G H 1-95 and 151-191 b G H 96-133 (calcd) TbGH-d TbGH-d (HoACI
4
28
17
4
29
21
4 4 9
36 29 22
8 IO 23
9
26
25
9 9
40 19
18
+
+
WAVELENGTH, nrn
3
5: Circular dichroism spectra of recombined bGH 1-95 and 151-191 and bGH 96-133 (pH 4). Recombined peptides (0.232 mg/mL) (-). Calculated peptides curve (- - -). TbGH-d treated with HoAC -). All solutions (0.231 mg/mL) (- - -). TbGH-d (0.22 mg/mL) were prepared in 0.01 M formate buffer, pH 4. Path length, 0.1 cm. The signal-to-noise ratio at 222 nm was 25:l and at 208 nm was 18:l.
following formula:
duplicated the absorption spectrum obtained when the individual peptides were in separate compartments of a quartz tandem cell in 0.1 N acetic acid. Except for some slight discrepancy in the region between 245 and 270 nm, the recombined peptides gave ultraviolet absorption spectra in 0.1 M sodium carbonate buffer, p H 9.5, similar to bGH (not shown). The circular dichroism spectra between 250 and 200 nm of bGH 1-95 and 151-191 and bGH 96-133 in separate compartments of a quartz tandem cell, and the peptides in 0.1 N acetic acid are shown in Figure 4. The C D spectrum of the recombined complex contains two troughs a t approximately 220 and 21 0 nm with ellipticities, [e], of - 14 300 and - 14 500. There is a crossover a t 202 nm. The circular dichroism spectrum of b G H 1-95 and 151-191 contains one trough a t approximately 220 nm with ellipticity, [e], of -10 600, and that of b G H 96-133 contains two troughs at 222 and 207 nm with ellipticity, [e], of - 12 400 and - 14 000. The C D spectra of the recombined complex at pH 4 (Figure 5 ) and p H 9 (Figure 6) are similar to TbGH-d treated with acetic acid and lower in ellipticity than the native TbGH-d spectra. The - - - line is a calculated curve obtained by the
136 38 174[e]bGH1-95andl51-19] ,h 174[~IbGH96-133.X where [e] is the molar ellipticity of the peptide C D spectrum, and 136, 38, and 174 are the number of amino acids respectively of bGH 1-95 and 151-191, b G H 96-133, and the recombined complex of these two peptides. Assuming no interaction between the two peptides, the calculated CD curve of thepeptidesismerelythesumofthebGH 1-95and 151-191 and bGH 96-1 33 individual spectrum characteristics a t the particular wavelength, X. The calculated curve is different in line shape from the recombined complex spectrum due to the contribution from the bGH 96-1 33 random coil spectrum. The secondary structural parameter calculation (Table 11) shows that TbGH-d contains a large amount of helix structure and almost no p sheet. The peptides of the recombined complex, TbGH-d (HoAC) and bGH 1-95 and 151-191 and bGH 96-133 (calcd) all have comparable amounts of helix and psheet structure. The corresponding fluorescence spectra at p H 9 and p H 4 (not shown) of the recombined peptides have a similar emission maximum a t 330-335 nm as that of TbGH-d, but the relative fluorescence intensity is higher. The calculated curve, taking into account the smaller proportion of tryptophan in the
FIGURE
a
Secondary structural parameters calculated by the method of
Chen et al. (1972).
(-..e
[%lcalcd,X =
+
554
BIOCHEMISTRY
flii
t s"'""l
155
l--.-I
HAKA. CIIEN, 4 N U SOXENBERG
I b 9 3 3 P d 51-15,
1
,
,
8
1
8
1
, ,
100
10
1
IOOC
DOSE (moles x
F I G U R E 7: Biological response of recombined peptides. The tibial width response, w, with standard error of the response have been plotted against the nanomolar dose on a log scale.2
complex than in bGH 1-95 and 15 I - 19 1 , is different from the recombined complex spectrum indicating a modification of the tryptophan fluorophore in the presence of b C H 96- 133. Biological Response. When bGH 96-1 33 was assayed a t doses (weight) approximately 20 times greater than the native hormone, the International Growth Standard, parallel responses vs. log dose were obtained (Figure 7). Thus, on a weight basis, b C H 96-133 had approximately 5% of the growth promoting potency of b C H ; 011 a molar basis this was about 1% of the growth promoting potency. The larger peptide, bGH 1-95 and 15 1 I9 1 , even a t maximal doses produced a response which barely exceeded 200 and its slope was distinctly more shallow than the bGH and bGH 96-1 33 slope of response vs. log dose. The growth promoting activity of the recombined equimolar amounts of the larger and smaller peptides was significantly greater than the growth promoting activity of bGH 96-1 33 with b C H 1-95 and I51 - I9 1 considered to have insignificant growth promoting activity. When both peptides were recombined in acetic acid. the resultant complex produced a slope of the dose--response curve not significantly different from that of bGH and bGH 96 -1 33. The molar potency estimate for the complex recombined in acetic acid was about 10% of the standard. When both peptides were recombined in urea, the slope of the dose-response curve for the complex was significantly steeper than the slopes of bGH and bGH 96-133. Under thc circumstances no potency estimate of this complex could be made. ~~
Discussion bGH 96- 133 retained significant capacity to increase tibial width (Table 111; see Supplementary Material Available paragraph a t end of paper) and produce weight gain in hypophysectomized rats (Yamasaki et al., 1970), but the biological activity of bGH 96-133 was markedly decreased compared with the parent bGH protein (Table 111; Yamasaki et al., 1970). However, the slope in a log dose-response plot was not significantly different for bGH 96-133 compared with bGH (Table 111). It should be noted that in the present study and more recent studies (Hara and Sonenberg, 1977) with 2 The bioassay data with the statistical evaluation were submitted to thc hcrutinq of the reviewers and will appear in the microfiche version. They m a l be obtained by interested readers by writing directly to the a u thorr
bGH 96-1 33, the slope of the log dose---responseplot is parallel to that of the International Growth Hormone Standard. This may be related io the larger doses (500 to 4500 kg per rat per 4 days) employed in these studies than the doses (30 to 600 pg per rat per 4 days) employed in earlier htudics (Yamasaki et al., 1970; Grkf and Li, 1974b; Li and G r i f , 1974; Grkf er al., 1976). In a sigmoidal dose-response curve. it may be that inadequate doses of bGH 96-133 with low activity were administered i n the earlier studies to detect parallelism with native growth hormone. Low biological activity in various small peptides in the region of residues 96 to 133 of grohth hormone was ilso observed in related experiments in other laboratories (BI; ke and Li. 1973: Chillemi et al.. 1972; Li'ang ei a!.. 19; 3 ) . The biological activity of undigested bGH decreased from 1 .O to 0.37 f 0.06 IU/mg when dissolved in 35% acetic acid for I8 h a t 4 "C (Sonenberg et al., 1972). The nature of the structural modification was not identified. The effects of urea on the biological activity of bGH and TbGH-x were tested before use as a dissociating agent with TbGH-x. Bioassay results showed that 8 IM urea did not reduce the growth promoting activity of bGH angd TbGH-x. Urea, 8 M, \vas shown to dissociate TbGH-x into three major components. One consisted of a peptide originating in residues 96- 133 of the bGH molecule (Wallis. 1973; Santorn6 et al.. 1973; G r i f and Li, 1974a). Another consisted of the amino-terminal position ( 1 95 of the sequence of bGH) and carboxyl-terminal position ( 1 5 1 -- !91 of the sequence of b G H ) which are linked by a disulfide bond. The sniiillest fraction consisted of residues 134--150of the bGH molecule. From these results, TbGH-x was formed by the cleavage o f ;irginylvaline. arginylalanine, and arginylserine peptide bonds occurring within the large disulfide loop of bGH. Previously Yamasaki et ai. (1970) showed two homogeneous components, peptide .A-1 and A-I I , were isolated from a single component of a tryptic digest of bGH (TbGH-d). TbGH-d had been obtained as the fraction eluted from a DEAE-cellulose column with 0.05 M ammonium bicarbonate buffer. For the present experiments, TbGH-x appeared in fractions eluted u.ith 0.03 M and/or 0.05 M ammonium bicarbonate buffer. In the present experiments. from the single component of the 0.05 M ammonium bicarbonate fraction (TbCH-x), we prepared three peptides using 8 M urea. .!\cctic acid. 5096. was also shown to dissociate TbGH-x into three components. These results show that TbGH-x may be different from TbGH-d and is possibly the c or e component ofTbGH (Yamasaki et al., 1970), or that the smallest peptide (bGH 134- 1 50) u a s lost during the purification procedure o f TbGH-d (Yam Previous11 it \vas presumed that the 16 000 molecular weight peptide originated in residues 1-95 and 134- I9 1 of the bGH molecule (Wallis. 1973: SantoniC et al.. 1973; G r i f and Li. 1974a), but the present results show that peptide A-I consisted of the residues 1-95 and I S 1 19 I of the sequence of bGH which are linked by a disulfide bond (Figure I ) . Based on the properties of the large and the small peptides under different conditions, there seems to be no obvious difference found i n peptides isolated by 50T0 acetic acid or 8 M urea. We attempted to recombine the three peptides, bGH 1-95 s n d 151-191. bCH 96- 133 and bCH 133-150. Equimolar amounts of these peptides were incubated in 1 .O M acetic acid. Only bGH 1-95 and I S 1 191 and bGH 96--133wererecombined and bGH 134- 1 5 0 behaved as another component. From these results and those of others (Lewis et al.. 1973; Singh et ai., 1974; Reagan et al., I975a.b) residues 134- 150 of the bCH molecule seem not to be necessary for the biological activity of growth hormone. ~
-
RECOMBINATION OF GROWTH HORMONE PEPTIDES
The recombination of two peptides was performed in 1.O N acetic acid because the association of bGH 96-133 was markedly dependent on the concentration of acetic acid employed as solvent. bGH 96-1 33 was found to exist as a monomeric form in acetic acid solutions, 1.O N or greater. Decrease in the concentration of acetic acid promoted self-association of bGH 96-133 and dimeric and trimeric forms were observed in 0.1 N acetic acid (Yamasaki and Kangawa, 1973). The C D spectra of the recombined complex are similar to TbGH-d treated with acetic acid, but have less ellipticity than TbGH-d. The calculated curve of the individual peptides is comparable in ellipticity but slightly different in line shape to the recombined complex. The C D spectrum of the recombined complex (pH 9) is similar to the recombined complex spectrum (pH 9.5) reported by Sonenberg and Beychok (1971). They showed that there is a significant difference in ellipticity between the spectrum of the recombined peptides and the tandem cell spectrum of bGH 1-95 and 151-191 and bGH 96-133 a t pH 9.5. The native TbGH-d is more helical and tighter structured than the complex. TbGH-d treated with 50% acetic acid could be looked upon as an intermediate structure in that its secondary structure resembles that of the recombined peptides, whereas its fluorophore characteristic resembles that of native TbGH-d. Fluorescence spectra indicate a conformational change upon recombination since the fluorescence intensity is reduced from the fluorescence predicted from the sum of the intensities of the individual peptides. This induced conformational change must be essential for the more complete expression of the growth promoting biological activity. The molecular organization seems to reflect biological activity. TbGH-d elicits a higher biological response than both the recombined peptide and TbGH-d (HoAC). The recombined complex is in turn higher than the sum of the effects of the individual peptides. It is noteworthy that the complex of bGH 1-95 and 15 1- 19 1 and bGH 96- 133 produced by recombination in urea was different than that obtained after recombination in acetic acid as indicated by different slopes of their dose-response curves. No spectroscopic comparisons of these two types of recombinations were made. CD spectroscopy does not predict the absolute determinants of biological activity. bGH contains 63% a helix and 3% 0 sheet; TbGH-d contains 29% a helix and 29% 0 sheet (Chen and Sonenberg, 1977), whereas their biological activities are of equivalent magnitude. Thus, molecular disorganization to a certain degree is perhaps allowed in growth hormone without destruction of biological activity. The present studies have not been able to identify the exact conformations of the individual peptides in their recombined state. The bGH structure predicted based on primary sequence (Chen and Sonenberg, 1977) shows that the bGH 96-133 segment has helical structure, which is probably related to the bGH 96- 133 helical state a t p H 4 as observed by CD. Therefore, the bGH 96-133 helical conformation is perhaps the active form of the bGH 96-133. When bGH 1-95 and 15119 1 and bGH 96- 133 were recombined in 1 N acetic acid, bGH 96-133 was already in a helical conformation which could be easily incorporated into the complex with some more conformational adjustments. bGH 1-95 and 151-191 contains a high percentage of 0structure which was not seen in bGH indicating that the structure of bGH 1-95 and 151-191 is different from the native bGH 1-95 and 15 1- 19 1 conformation originally in bGH. The lack of conformational flexibility of this peptide indicates that it is frozen in the high 0 sheet structure state which is more difficult to revert back to native peptide in bGH. This may explain why the recombined complex is higher in 0 sheet content, lower in ellipticity and lower
VOL.
17,
NO.
3, 1978
555
in biological activity than native TbGH-d. Studies of the structure and biological activity of the two individual peptides alone and recombined suggest that these two peptides are both essential for the more complete expression of growth hormone activity. Acknowledgments The authors are grateful to Dr. Gary Litman for his assistance in determinations of amino acid composition. Supplementary Material Available The biological response of recombined peptides (Table 111) (1 page). Ordering information is given on any current mast-
head page. References Blake, J., and Li, C. H. (l973), Int. J . Peptide Protein Res. 5, 123. Chen, C. H., and Sonenberg, M. (1977), Biochemistry 16, 21 10. Chen, T. H., Yang, J. T., and Martinez, H. M . (1972), Biochemistry 11, 41 20. Chillemi, F., and Pecile, A. (1971), Experientia 27, 385. Chillemi, F., Aiello, A., and Pecile, A. (1972), Nature (London), New Biol. 238, 243. Davis, B. J. (1964), Ann. N.Y. Acad. Sci. 121, 404. Dellacha, J. M., and Sonenberg, M. (1964), J . Biol. Chem. 239, 1515. Free, C. A., and Sonenberg, M. (1966), J . Biol. Chem. 241, 5076. Geschwind, I. I., and Li, C. H. (1955), in Hypophyseal Growth Hormone, Nature and Actions, Smith, R. W., Gaebler, 0. H., and Long, C. N . H., Ed., New York, N.Y., McGrawHill, pp 25-58. GrBf, L., and Li, C. H . (1974a), Biochem. Biophys. Res. Commun. 56, 168. GrBf, L., and Li, C. H . (1974b), Biochemistry 13, 5408. Grbf, L., Barat, E., BorvendCg, J., Hermann, I., and Pattny, A. (l976), Eur. J. Biochem. 64, 323. Gray, W. R., and Hartley, B. S. (1963), Biochem. J . 89, 59P. Hara, K., and Sonenberg, M. (1977), Biochim. Biophys. Acta 492, 95. Levine, L. S., Sonenberg, M., and New, M. I. (1973), J . Clin. Endocrin. Metab. 37, 607. Lewis, U. J., Singh, R. N. P., Lindsay, T. T., Seavey, B. K., and Lambert, T. H . (1973), in Advances in Human Growth Hormone Research, Raiti, S., Ed., Washington, D.C., D H E W Publication No. ( N I H ) 74-612, p 349. Lewis, U. J., Pence, S. J., Singh, R. N. P., and Van der Laan, W. P. (1975), Biochem. Biophys. Res. Commun. 67, 617. Li, C. H., and Bewley, T. A. (1976), Proc. Natl. Acad. Sci. U.S.A.73, 1476. Li, C. H., and GrBf, L. (1974), Proc. Natl. Acad. Sci. U.S.A. 71, 1197. Li, C. H., Hayashida, T., Doneen, B. A., and Rao, J. (1976), Proc. Natl. Acad. Sci. U.S.A. 73, 3463. Li, C. H., Bewley, T. A., Blake, J., and Hayashida, T. (1977), Proc. Natl. Acad. Sci. U.S.A.74, 1016. Lowry, 0.H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951), J . Biol. Chem. 193, 265. Nadler, A. C., Sonenberg, M., New, M. I., and Free, C. A. (1967), Metab., Clin. Exp. 16, 830. Ornstein, L. (1964), Ann. N . Y. Acad. Sci. 121, 321. Reagan, C. R., Mills, J. B., Kostyo, J. L., and Wilhelmi, A. E.
556
BIOCHEMISTRY
( 1 975a), Endocrinology 96, 625. Reagan, C. R., Mills, J. B., Kostyo, J . L., and Wilhelmi, A . E. (1975b), Proc. Natl. Acad. Sci. U.S.A. 72, 1684. Reisfeld, R . A., Lewis, U. J.. and Williams, D. E. (l962), 'Yarure (London) 195, 28 1, Santomi, J. A , , Dellacha, J. M., Paladini, A. C., Pence, C., Biscoglio, M. J.. Daurat, S. T.. Poskeno. E., and Wolfenstein, C . E. M. (1973), Eur. J . Biochem. 37, 164. Shapiro, A . L., Vinuela, E., and Maizel, Jr., J . V. (1967), Biochem. Biophys. Res. C'otnmun. 28. 8 15. Singh, R. N. P., Seavey, B. K., Rice, V. P., Lindsay, T. T., and Lewis, U. J . (1974), Endocrinology 94, 883. Sonenberg, M . (1971), Proc. Natl. Acad. Sci. U.S.A. 68,
1051.
Sonenberg, M., and Beychok. S. (1971), Biochim. Biophys. Acta 229, 88. Sonenberg, M., Free, C. A., Dellacha, J . M.. Bonadonna, G., Haymowitz, A . . and Nadler, A. C. ( 1 965), Metab., Clin. E s p . 14, 1189. Sonenberg, M., Kikutani, M., Free, C. A,, Kadler, A . C., and Dellacha, J. M. (1968), Ann. N.Y. Acad. Sci. 148, 532. Sonenberg, M., Yamasaki, N., Kikutani, M.,Swislocki, N . I., Levine, L., and New, M.I . , (1972), Excerpta Med. Found. Int. Congr. Ser., No. 244. 75.
Sonenberg, M., Levine, L. S., New, 41.I., and Saxena, B. 9. (1973), Advances in Human Growth Hormone Research, Washington, D.C., D H E W Publication N o . (NIH) 74-61 2, p 467. Spackman, D . H., Stein. W . H., and Moore, S. (1958), Anal. Chem. 30, 1190. Suislocki, N.I., Sonenberg, M., and Yamasaki, N . (1970), Endocrinology 87, 900. Wallis, M.(1 973), FEBS Lett. 35, 1 1. Wang, S. S., Yang, C. C., Kulesha, I. D., Sonenberg, M., and Merrifield, R. B. ( I 9741, Int. J . Peptide Protein Res. 6, 103. Weber, K., and Osborn, M. (1969), J . Biol. Chem. 244, 4406. Woods, K. R.. and Wang, K.-T. (1967), Biochim. Biophys. Ai~ta133, 369. Yamasaki, N.,and Kangawa, K. (1973), Agric. Biol. Chem. ( T o k y o ) 37,2637. Yamasaki, K.,Kikutani, M., and Sonenberg, M. (1970), Biochemistry 9, 1107. Yamasaki. N , , Kangawa. K., Kobayashi, S., Kikutani, M., and Sonenberg, M . (1972). J . Biol. Chem. 247, 3874. Yamasaki, N . , Simanaka, J . , and Sonenberg, M. (1975), J . Biol. Chem. 250, 2510.
Biosynthesis of the Macrolide Antibiotic Chlorothricin: Basic Building Blocks? Rainer Holzbach, Hermann Pape,? Derek Hook, Ernest F. Kreutzer, Ching-jer Chang, and Heinz G . floss*^§
ABSTRACT: The biosynthesis of chlorothricin (I), a macrolide antibiotic isolated from Streptomyces antibioticus TU 99, has been studied by feeding experiments with I4C- and 3H-labeled precursors. Acetate and propionate, but not methionine and mevalonate, were incorporated into the macrocyclic aglycone of the antibiotic. Glucose and the various carbon atoms of tyrosine, except the carboxyl carbon, also contributed label to
c
hlorothricin (I) is a novel macrolide antibiotic which was first isolated from Streptomyces antibioticus strain TU 99 (Keller-Schierlein et al., 1969a). Its antibiotic activity against gram-positive bacteria on synthetic, but not on complex media, is due to an inhibition of the anaplerotic C02 fixation catalyzed by pyruvate carboxylase (Schindler & Zahner, 1972). Further mode of action studies have also shown an interaction between chlorothricin and the membrane phospholipids of Bacillus subtilis (Pache & Chapman, 1972). In its binding to pyruvate carboxylase from B. stearothermophilus, chlorothricin anFrom the Institut fur Biologie, Universitit Tubingen, Tiibingen, West Germany, and the Department of Medicinal Chemistry and Pharmacognosy, Purdue University, West Lafayette, Indiana 47907. Receiced J u l y I . 1977; rerised manuscript receiced November I O , 1977. This work was supported by U S . Public Health Service Research Grant A I 11728 from the National Institute for Allergy and Infectious Diseases. Present address: Institut fur Mikrobiologie. Universitat Miinster. Munster, West Germany. § Address correspondence to this author a t Purdue University.
the aglycone. Glucose also seems to be a specific precursor of the 2-deoxyrhamnose moiety, probably via a process involving a hydrogen shift from C-4 to C-6 of the hexose. The substituted 6-methylsalicylic acid moiety seems to be derived from acetate and one O-methyl group provided by methionine; shikimic acid is not incorporated.
tagonizes the activating effect of acetyl-coenzyme A on this enzyme (Schindler & Zahner, 1973). Different inhibition
+
I : x-c1 11: X - H I11 : X
-
Br
C I978 American Chemical Society
